Deep-brain stimulation (DBS)—which delivers electrical pulses by wires directly to select brain areas—has been shown to calm Parkinson's tremors, and is being explored as a way to alleviate depression and relieve obsessive-compulsive behaviors. Could it improve memory, too? Perhaps, suggests a February 9 New England Journal of Medicine paper by Itzhak Fried, University of California, Los Angeles, and colleagues. They report that epilepsy patients who receive electrical stimulation to the entorhinal cortex during learning improve their performance in a virtual reality spatial memory task. This type of memory is one of the earliest affected in Alzheimer's disease (AD) and the results could mean DBS might help. The work made the pages of The New York Times. Meanwhile, another virtual reality navigation task may distinguish AD from frontotemporal lobar degeneration, suggests a paper in the February 8 Journal of Neuroscience by Laure Rondi-Reig and colleagues at Pierre et Marie Curie Paris University in France. The authors suggest that the test could be used as part of a cognitive and psychological test battery to improve accuracy of AD diagnosis.

Deep-brain stimulation has been used for over a decade to treat Parkinson's disease, and more recently has been applied to other cognitive and psychological disorders caused by over- or underactive brain activity (see ARF related news story). In 2008, DBS to the hypothalamus jogged the memory of a man undergoing surgery for morbid obesity and improved his performance on a verbal memory test (Hamani et al., 2008). In later tests of people with AD, DBS was shown to be safe and boost metabolism in the temporal and parietal areas. While no cognitive benefits were suggested in this small trial, it seemed that participants' cognitive scores declined more slowly than would have been expected (see ARF related news story).

While it is not clear exactly how DBS works, enhanced neurogenesis has been suggested as one potential mechanism (see ARF related news story). In fact, at the 2011 Society for Neuroscience meeting in Washington, DC, researchers from Andres Lozano’s lab at the University of Toronto, Canada, reported that, in a small clinical trial, DBS slowed hippocampal atrophy on average, and even increased hippocampal volume in two patients.

In the NEJM paper, first author Nanthia Suthana and colleagues recruited seven epilepsy patients who were having electrodes implanted to identify seizure zones for possible surgical removal. These volunteers had electrodes placed in several areas, including the hippocampal area, entorhinal cortex, or both. These electrodes delivered electric pulses specifically to these areas.

After electrodes were in place, participants played a virtual game from laptops in their hospital beds. Using a joystick, patients "drove" a taxi around a virtual town, picking up passengers and delivering them to one of six grocery stores. Volunteers completed this sequence of stores three times, receiving electrical pulses as they drove to either even- or odd-numbered stores. Since the volunteers couldn't feel anything, they could not tell which routes were accompanied by pulses and so served as their own controls. On the fourth try, patients completed one final set of trips without any stimulation. Researchers measured the excess path length (actual distance driven minus the shortest possible path) to each store and the time it took to get there.

Pulses delivered to the entorhinal cortex during learning led volunteers to find the most efficient routes (shortest path lengths and quickest driving times). Patients showed the benefit only on routes to the three stores for which they received stimulation during learning. Trials with no stimulation or involving pulses to the hippocampus did not show the same improved outcome.

"Stimulating this particular area [the entorhinal cortex] seems to enhance memory," Suthana told ARF. It is not clear whether the stimulation improves memory formation, or retrieval, or both. The team also does not have a handle on how long these effects last or what mechanism underlies DBS improvements, so whether or not this is a viable option for people with memory disorders remains to be seen, she said. "It is an exciting possibility."

Practicality of such a treatment might depend on the type of memory loss, added Sandra Black, University of Toronto, Canada, who authored an NEJM editorial on the work (Black, 2012). DBS might be more applicable to stable memory disorders—such as memory loss from stroke—than progressive ones such as Alzheimer's, she wrote, suggesting DBS may not be able to keep up with an ongoing "siege" of toxic amyloid and tau pathology in AD, at least not for very long. However, "if we can figure out some way to stall the disease at an earlier stage, then maybe [DBS] would be very helpful for gaining back some of that [lost memory] capacity," Black suggested. Such basic research will explore the therapeutic possibilities of DBS, and may prepare the field to use future technological advances that stimulate brain areas in a less risky and invasive way.

Next, Suthana's team will use the same setup with a different set of patients to test whether stimulation during recall alone, or during both learning and recall, confers similar benefits.

Virtual reality tasks may offer additional clues to AD diagnosis, suggests the paper by Rondi-Reig and colleagues. The team reports that a spatiotemporal test involving virtual navigation through a maze differentiates patients with Alzheimer's disease from those with frontotemporal lobar degeneration (FTLD). Modeling the Starmaze navigation test, which is meant to assess hippocampal function in rodents, first author Virginie Bellassen and colleagues developed a non-verbal virtual reality game in which people use a joystick to maneuver through a pentagonal maze with five arms radiating from its corners. Mountains, radio antennae, and forests served as landmarks. Over 11 learning trials, participants learned the route to a prize, always hidden in the same spot. On the final test trial, they had to remember the sequence of turns through the virtual maze to find the prize without the aid of landmarks, then trace out the route on a piece of paper—both considered temporal tasks. Researchers also asked participants to point out where the landmarks had been—a spatial task.

The team tested four groups of people—63 who were cognitively normal (divided into three age groups, 20-39, 40-59, 60-80), 16 who had clinically diagnosed AD, 14 with amnestic mild cognitive impairment (aMCI), and 11 with FTLD. While the volunteers with FTLD, aMCI, and AD all had difficulty with the spatial aspect of the test, only people with AD and aMCI faltered on the temporal tasks. When the temporal test was combined with other standard neuropsychological tests, the navigation test brought sensitivity and specificity of AD detection to 100 percent. "If adapted in the memory center, it could help with the tests that are already used to diagnose [AD], and avoid false positives and false negatives," Rondi-Reig told ARF.

Preliminary results also suggest that the temporal tasks may also pick up very early signs of AD, the authors wrote. Those who converted from aMCI to AD within three years had some of the lowest temporal scores in the aMCI group. In addition, two healthy controls in the 60-80 age range with the lowest scores in their group started having episodic memory deficits 18 months after the test.

Though the test's length needs to be shortened to be clinically applicable, it could be used in a cognitive test battery, Rondi-Reig told ARF. All it requires is a computer, a joystick, and the game, so it should not be too expensive or require special equipment. She and her team next plan to use this test along with functional neuroimaging and diffusion tensor imaging to see if there are changes to patients' neural networks or neural connections that underlie temporal order memory.

The navigation task taps into an early, and often overlooked, complaint of patients with early AD, said Charles Duffy, University of Rochester Medical Center, New York, who has done his own research into navigational deficits in people with AD (Cushman et al., 2008). One of the patients' first problems is that they can no longer find their way to the grocery store or a relative's house, though they have been traveling that route for many years. "That's what's truly debilitating about the early stages of this disease," he said, adding that it robs patients of their independence. Alzheimer's presents differently for everyone, so it makes sense that such a navigational test should fit in with a larger battery of tests that address various aspects of potential memory deficits, he said.—Gwyneth Dickey Zakaib


  1. The work from the group at UCLA is exciting and could have important implications for patients with memory disturbances. Just as deep-brain stimulation has been used to modify the activity of brain areas involved in movement in over 80,000 patients with Parkinson's disease, here we now have an example of using stimulation to modify the performance of brain areas involved in memory. This work shows that the circuits in the brain that are responsible for memory are reachable, and that their activity can be adjusted by electrical stimulation. Although the number of patients was small and they applied it to patients being treated for epilepsy rather than memory disturbances, it lends support to the concept that stimulating specific brain areas to improve memory could potentially be useful therapeutically.

    Our own group has applied deep-brain stimulation in a small number of patients with Alzheimer's disease at another location along the same memory circuit, the fornix (Laxton et al., 2010). We are building upon our previous work and that of the group at UCLA, and hope to launch a Phase 2 trial of DBS for mild Alzheimer's disease in the near future.


    . A phase I trial of deep brain stimulation of memory circuits in Alzheimer's disease. Ann Neurol. 2010 Oct;68(4):521-34. PubMed.

    View all comments by Andres M. Lozano

Make a Comment

To make a comment you must login or register.


News Citations

  1. Deep-Brain Stimulation: Decade of Surgical Relief, Not Just for PD
  2. DBS Update: Attempting to Stimulate Memory in Alzheimer’s
  3. Does Deep-Brain Stimulation Spark Neurogenesis, Enhance Learning?

Paper Citations

  1. . Memory enhancement induced by hypothalamic/fornix deep brain stimulation. Ann Neurol. 2008 Jan;63(1):119-23. PubMed.
  2. . Brain stimulation, learning, and memory. N Engl J Med. 2012 Feb 9;366(6):563-5. PubMed.
  3. . Detecting navigational deficits in cognitive aging and Alzheimer disease using virtual reality. Neurology. 2008 Sep 16;71(12):888-95. PubMed.

External Citations

  1. The New York Times

Further Reading

Primary Papers

  1. . Memory enhancement and deep-brain stimulation of the entorhinal area. N Engl J Med. 2012 Feb 9;366(6):502-10. PubMed.
  2. . Brain stimulation, learning, and memory. N Engl J Med. 2012 Feb 9;366(6):563-5. PubMed.
  3. . Temporal order memory assessed during spatiotemporal navigation as a behavioral cognitive marker for differential Alzheimer's disease diagnosis. J Neurosci. 2012 Feb 8;32(6):1942-52. PubMed.